From e1a24f6250f5e498e524ce14a5acbed5fc5c2a91 Mon Sep 17 00:00:00 2001 From: Anders Betts Date: Tue, 22 Sep 2026 19:52:19 +0200 Subject: reports: income statement skips imported Stäng closings MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-Authored-By: Claude Opus 5.5 --- docs/DECISIONS.md | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'docs/DECISIONS.md') diff --git a/docs/DECISIONS.md b/docs/DECISIONS.md index a0c613b..a109bda 100644 --- a/docs/DECISIONS.md +++ b/docs/DECISIONS.md @@ -333,7 +333,15 @@ kept verbatim from the STATE.md they were pruned from (2026-09-21). statement (Kapitas 2022-2026). **Decided 2026-09-19: no importer change.** Locked years stay locked and the source's closings stay in the books; the årsredovisning export flags incomplete jämförelsetal for those years and - points to the previous year's annual report. + points to the previous year's annual report. **Amended 2026-09-22**: the + income statement (and therefore INK2/SRU and the TUI resultatrapport) now + skips the same "Stäng ..." vouchers, so a year being declared shows its + real figures; the balance sheet keeps them (the result sits in 2099 and + must not be counted twice). The source's own #IB/#UB corrections that the + vouchers do not reproduce (e.g. Makandra's FY 2022/2023 result transfer + is 17 857.63 kr short of that year's P&L) remain a data divergence: the + derived balance sheet is off by that amount until a correction is booked + in the current year. 11. ~~SIE import only into an empty fiscal year; consider broader import.~~ Chronological multi-year import works (CRLF, `#RAR 0`, zero rows, `#IB` rule handled); each year must still target an empty fiscal year. Note: -- cgit v1.3